The Agent Skills Directory

[SAFE]: The skill is authored by 1Password and provides a secure method for agents to interact with 1Password Developer Environments through the official 1password-mcp server.
[EXTERNAL_DOWNLOADS]: The skill refers to official 1Password resources, including the 1password-mcp tool and the 1Password/1password-codex-plugin repository on GitHub, for its setup and operation.
[COMMAND_EXECUTION]: The instructions involve configuring and running the 1password-mcp local command. This execution is part of the primary, intended functionality of the skill to bridge the AI agent with the local 1Password environment.
[CREDENTIALS_UNSAFE]: The skill implements robust security instructions to prevent secret exposure. It explicitly directs the agent to only return variable names (never values), avoid logging secrets, and refrain from reading mounted .env files, ensuring sensitive data remains protected within the 1Password vault.

1password-environments